Extreme Quality on current high end systems? [Archive]

View Full Version : Extreme Quality on current high end systems?

Caldaini

05-10-2007, 03:20 AM

I am about to return to playing the game after 2 years. I'm getting a new PC with Core 2 Duo E6700, 768MB 8800 GTX, and 2 GB of DDR2. Will I be able to run the game at a playable frame rate yet? Two years ago nV news (<a href="http://www.nvnews.net/articles/everquest_2_graphics/index.shtml" target="_blank" rel="nofollow">http://www.nvnews.net/articles/ever...ics/index.shtml</a>) benchmarked the game with various settings and they had a top of the line PC back then and could only get 7 fps at max detail. Is it still running that poorly? Two years later, hardware has jumped leaps and bounds. By now, my hardware will have been considered 'futuristic' for 2004 standards would it not? Don't say it is because the engine is too advanced for the hardware. This is a 2004 game. Oblivion is a far better looking game than EQ2 at Extreme Quality, and it runs very nicely on current hardware. If anything, it's just a poorly optimized engine. Is anyone here able to run it at max? If so, what are your system specs?

Saavedra

05-10-2007, 03:41 AM

I'm running a core duo 2.6 with 2 gigs of ram and an 8800gts 640meg nvidia card and can run in extreme quality pretty well depending on the zone. I wouldnt recommend it in say Qeynos Harbor, but just about any overland zone runs pretty well. Lately I've settled for a balance of a tweaked version of Very High Quality, with a few settings changed for most grouping.

Mythal_EQ2

05-10-2007, 04:33 AM

You should be able to run EQ2 on max everything with that system, with the exception of Shadows. I have found that, even on my Quad-core system, enabling shadows is a significant performance hit.If I'm not mistaken SoE is working on a new shadow system on Test right now, so that might improve in the near future, but for now your best bet is to keep them turned off.For your eyes' sake, you might want to turn off particle effects if you're raiding too <img src="/smilies/69934afc394145350659cd7add244ca9.gif" border="0" alt="SMILEY" /> Not for any performance issues, because the 8800's have no trouble handling them, but because you'll probably go blind from all the colors.

Savanja

05-10-2007, 04:49 AM

I have shadows turned off, something that was suggested to me by a guildmate to ease performance issues when running the higher quality settings. It does help quite a bit. My set up isn't as good as yours, but I still run in high quality most of the time and drop it when I need to, to balanced for smooth performance with no stuttering in areas that tend to be mean on the system.

Serso

05-10-2007, 05:55 AM

<cite>Caldaini wrote:</cite><blockquote>I am about to return to playing the game after 2 years. I'm getting a new PC with Core 2 Duo, 768MB 8800 GTX, and 2 GB of DDR2. Will I be able to run the game at a playable frame rate yet? Two years ago nV news (<a href="http://www.nvnews.net/articles/everquest_2_graphics/index.shtml" target="_blank" rel="nofollow">http://www.nvnews.net/articles/ever...ics/index.shtml</a>) benchmarked the game with various settings and they had a top of the line PC back then and could only get 7 fps at max detail. Is it still running that poorly? Two years later, hardware has jumped leaps and bounds. By now, my hardware will have been considered 'futuristic' for 2004 standards would it not? Don't say it is because the engine is too advanced for the hardware. This is a 2004 game. Oblivion is a far better looking game than EQ2 at Extreme Quality, and it runs very nicely on current hardware. If anything, it's just a poorly optimized engine. Is anyone here able to run it at max? If so, what are your system specs? </blockquote> This probably originates from the fact that Elder Scrolls IV: Oblivion was a multi-platform title -- graphical code usually goes through extensive evaluation by many tempered (properly synchronized for coordination; well-versed [in terms of visuals implementation]) compilers and coders known for efficiency optimization. When developments have that kind of backing, I suppose it would be easier to enlist the necessary utilities for these types of measures. Televised advertisement, I feel (along with the backing of many others I'm sure), is an accommodation beckoning enacting. - Seliri D. of Nagafen

Norrsken

05-10-2007, 06:10 AM

<cite>Serso wrote:</cite><blockquote><cite>Caldaini wrote:</cite><blockquote>I am about to return to playing the game after 2 years. I'm getting a new PC with Core 2 Duo, 768MB 8800 GTX, and 2 GB of DDR2. Will I be able to run the game at a playable frame rate yet? Two years ago nV news (<a href="http://www.nvnews.net/articles/everquest_2_graphics/index.shtml" target="_blank" rel="nofollow">http://www.nvnews.net/articles/ever...ics/index.shtml</a>) benchmarked the game with various settings and they had a top of the line PC back then and could only get 7 fps at max detail. Is it still running that poorly? Two years later, hardware has jumped leaps and bounds. By now, my hardware will have been considered 'futuristic' for 2004 standards would it not? Don't say it is because the engine is too advanced for the hardware. This is a 2004 game. Oblivion is a far better looking game than EQ2 at Extreme Quality, and it runs very nicely on current hardware. If anything, it's just a poorly optimized engine. Is anyone here able to run it at max? If so, what are your system specs? </blockquote> This probably originates from the fact that Elder Scrolls IV: Oblivion was a multi-platform title -- graphical code usually goes through extensive evaluation by many tempered (properly synchronized for coordination; well-versed [in terms of visuals implementation]) compilers and coders known for efficiency optimization. When developments have that kind of backing, I suppose it would be easier to enlist the necessary utilities for these types of measures. Televised advertisement, I feel (along with the backing of many others I'm sure), is an accommodation beckoning enacting. - Seliri D. of Nagafen </blockquote>Or I would say completely different types of games. The EQ2 engine handles a lot of stuff that oblivion does not. Such as other players. And lots of them.

Tallest

05-10-2007, 06:20 AM

I'm using a 2.4 Core 2 Duo and an 8800GTS and, as others have posted, only shadows cause problems (especially in Enchanted Lands... the individual shadows of all those wings :shudders<img src="/smilies/3b63d1616c5dfcf29f8a7a031aaa7cad.gif" border="0" alt="SMILEY" />

Caldaini

05-10-2007, 08:13 AM

Norrsken

05-10-2007, 08:30 AM

<cite>Caldaini wrote:</cite><blockquote>Ulvhamne@Nagafen wrote: <blockquote><cite>Serso wrote:</cite><blockquote><cite>Caldaini wrote:</cite><blockquote>I am about to return to playing the game after 2 years. I'm getting a new PC with Core 2 Duo, 768MB 8800 GTX, and 2 GB of DDR2. Will I be able to run the game at a playable frame rate yet? Two years ago nV news (<a href="http://www.nvnews.net/articles/everquest_2_graphics/index.shtml" target="_blank" rel="nofollow">http://www.nvnews.net/articles/ever...ics/index.shtml</a>) benchmarked the game with various settings and they had a top of the line PC back then and could only get 7 fps at max detail. Is it still running that poorly? Two years later, hardware has jumped leaps and bounds. By now, my hardware will have been considered 'futuristic' for 2004 standards would it not? Don't say it is because the engine is too advanced for the hardware. This is a 2004 game. Oblivion is a far better looking game than EQ2 at Extreme Quality, and it runs very nicely on current hardware. If anything, it's just a poorly optimized engine. Is anyone here able to run it at max? If so, what are your system specs? </blockquote> This probably originates from the fact that Elder Scrolls IV: Oblivion was a multi-platform title -- graphical code usually goes through extensive evaluation by many tempered (properly synchronized for coordination; well-versed [in terms of visuals implementation]) compilers and coders known for efficiency optimization. When developments have that kind of backing, I suppose it would be easier to enlist the necessary utilities for these types of measures. Televised advertisement, I feel (along with the backing of many others I'm sure), is an accommodation beckoning enacting. - Seliri D. of Nagafen </blockquote>Or I would say completely different types of games. The EQ2 engine handles a lot of stuff that oblivion does not. Such as other players. And lots of them. </blockquote> The 7 fps that nV tested was at Isle of Refuge which had barely anyone. Even at lower detail settings the framerates weren't all that impressive for the hardware they had at the time. And if movie clips are anything to go by, even Crysis is looking to be smoother than EQ2, though I won't give them the benefit of the doubt just yet -- after all, SOE's official EQ2 movies are running very smoothly (what system were they using?). So I guess I can turn everything on except shadows (completely off), including the cloth simulation? I'll be happy with 30 fps. Oh and, how does Lavastorm run? That was the worst zone back when I was playing. Had to decrease it to minimum and was still getting less than 10 fps, even with no players around. </blockquote>You missed my point. The oblivion engines pretty much sole purpose is to look pretty. It doesnt have to bother much with network lag and all other kind of problems htat MMOs deal with. The engine is built differently, to different specs. THAT is why it is as it is. I am damned sure that oblivion can be outdone completely by a tech demo as far as prettyness. Tech demos doesnt have a high entertainment value though, other than the first few times youw atch them. ITs all depending on the spec with which you have to comply when you make your engine. Even though you may not be USING heavy networking, the code for it is still there, and it is doing its work anyways. And that particluar code took time to write. Time that a less network heavy game put on graphics.

Serso

05-10-2007, 08:59 AM

Ulvhamne@Nagafen wrote: <blockquote><cite>Serso wrote:</cite><blockquote><cite>Caldaini wrote:</cite><blockquote>I am about to return to playing the game after 2 years. I'm getting a new PC with Core 2 Duo, 768MB 8800 GTX, and 2 GB of DDR2. Will I be able to run the game at a playable frame rate yet? Two years ago nV news (<a href="http://www.nvnews.net/articles/everquest_2_graphics/index.shtml" target="_blank" rel="nofollow">http://www.nvnews.net/articles/ever...ics/index.shtml</a>) benchmarked the game with various settings and they had a top of the line PC back then and could only get 7 fps at max detail. Is it still running that poorly? Two years later, hardware has jumped leaps and bounds. By now, my hardware will have been considered 'futuristic' for 2004 standards would it not? Don't say it is because the engine is too advanced for the hardware. This is a 2004 game. Oblivion is a far better looking game than EQ2 at Extreme Quality, and it runs very nicely on current hardware. If anything, it's just a poorly optimized engine. Is anyone here able to run it at max? If so, what are your system specs? </blockquote> This probably originates from the fact that Elder Scrolls IV: Oblivion was a multi-platform title -- graphical code usually goes through extensive evaluation by many tempered (properly synchronized for coordination; well-versed [in terms of visuals implementation]) compilers and coders known for efficiency optimization. When developments have that kind of backing, I suppose it would be easier to enlist the necessary utilities for these types of measures. Televised advertisement, I feel (along with the backing of many others I'm sure), is an accommodation beckoning enacting. - Seliri D. of Nagafen </blockquote>Or I would say completely different types of games. The EQ2 engine handles a lot of stuff that oblivion does not. Such as other players. And lots of them. </blockquote> Yes, that would be a factor as well, but the basic algorithms for graphical dispersion permit maintained, smooth framerates within challenging brands that are able to recognize a baseline quality above that granted in EverQuest II. I'm not completely positive, but there emerges no emanant idea on the terms of renovations given to the rendering model of EverQuest II. <a href="http://www.anandtech.com/showdoc.aspx?i=2477&p=4" target="_blank" rel="nofollow">Here</a>, in a July 22nd, 2005 Anandtech analysis of GPUs, the ATI x850 XT in Crossfire is mentioned to receive 9-12 FPS while at a 1600x1200 resolution in "Ultra High/Extreme"* graphics settings - in "Very High" configurations, it receives a 40.8 FPS mark with a 6800 Ultra holding at 37 FPS. In <a href="http://www.anandtech.com/showdoc.aspx?i=2556&p=5" target="_blank" rel="nofollow">this</a> link (dating October 7th, 2005), the 6800 GT scores minutely lower than the 6800 Ultra did with the same resolution at "Very High" visual optioning (exampling general consistency). Where the paradigm shift occurs is when you consider nV News' <a href="http://www.nvnews.net/articles/everquest_2_graphics/index.shtml" target="_blank" rel="nofollow">sampling</a> (of January 8th, 2005). Given the fact that Anandtech doesn't provide FPS scenario screenshots, we can clearly comprehend, in all cases, that nV News benchmarking is done under a vast range of variability (all with the bells and whistles of 4X anti-aliasing and 8X anisotropic filtering [albeit, with the lesser resolution of 1028x768]). In these cases, the 6800 Ultra is defined to produce only 7 FPS at a resolution of 1028x768. Comparatively, the 6800 GS is outclassed by the 6800 Ultra in miniscule margins, except in core clock ratios (referring to an Anandtech 6800 <a href="http://images.anandtech.com/reviews/video/NVIDIA/6800GS/nv4xcomp.png" target="_blank" rel="nofollow">chart</a> found in a November 7th, 2005 <a href="http://www.anandtech.com/showdoc.aspx?i=2593&p=1" target="_blank" rel="nofollow">review</a>). Just contemplating the general level of performance cliff-diving when it comes to the handling of high-tier visual features (i.e. noting the aforementioned FPS drop from "Very High" to "Ultra High" at 1600x1200, even with a more robust x850 XT graphics solution), you can see how under-performing the systems behind the EverQuest II virtual realm are, especially when you note the correlatively high FPS average of 21.2 with Elder Scrolls IV: Oblivion graphics maximized to their highest degree while utilizing a 6800 GS at a 1280x1024 resolution (unlike for EverQuest II where "Ultra High" structuring generated massive FPS drop-off, let alone the fact that the score in Oblivion is able to surpass that of the 6800 Ultra with EverQuest II optioning at "Ultra High" by 14 points [if focusing on "Town" framerates from ES4:O], on top of having a resolution higher than 1028x76<img src="/smilies/b2eb59423fbf5fa39342041237025880.gif" border="0" alt="SMILEY" />. These ES4:O statistics came from an April 26th, 2006 Anandtech <a href="http://www.anandtech.com/showdoc.aspx?i=2746&p=4" target="_blank" rel="nofollow">analysis</a> done with a multitude of GPUs on Oblivion. Well, given the examination, I can only emphasize how insignificant player model coverage is in the nV News citation. Despite EQ2 being a MMORPG, I'm pretty sure that any tests made with ES4:O couldn't have been disqualified for equitable comparison when less than two models were available on-screen in evaluations showcasing super massively, infinitesimal FPS scores. If anything, having some models in view on a MMORPG when contrasting with a single-player RPG detracts from leeway had with the MMORPG -- depth in model articulation is something had on a lesser scale in a MMORPG than with that of an RPG revolving around an independent campaign. Something I'd like to note is the relative marks of cards in ES4:O to that of those surrounding results of the 6800 Ultra/GS in EverQuest II -- with the 7800 GTX scoring about 1.7 FPS lower than the x850 XT in EverQuest II at 1600x1200, the 7800 GTX benchmarked 49.4 FPS comparative to the prior mentioned 6800 GS framerate of 21.2. That's approximately an addition of 32 frames had at a 1280x1024 resolution. The x850 XT, said to get only 9-12 FPS at "Ultra High" configurations within EQ2 at a resolution of 1600x1200 doesn't make any large spacing from its lesser, nVidian 6800 Ultra predecessor when configs are clocked in at the "Ultra-High" mods. The capacity for such (spacing of new and old models) signifies a more efficient and proficient usage of processing capability -- something EverQuest II just doesn't show when portrayed among newer releases. Look also to the FPS leap when opting for 6800 GSes in SLI for ES4:O -- quite a decent modification given how sharply rates can turn sour with EQII once you deck on the high-tier visual amplifiers (21.2 to 33.1). Allocations for contemporary integration of sequences designed to appropriate key facets [enabling modern 3D renders ever-more grandiose feature-sets] are things introduced in later model cards as the generations progress -- to see the kind of jumps made with SLI 6800 GSes for ES4:O exhibit the fact that even such an aged card could manage values for a fresh hit it couldn't muster for an old one -- a signal that would logically be perceived as poor configuration. The amount of benefits seen from driver updates do tally to something, but commonly, hardware architecture stands as the core arbitrator. If you peek at the EQII SLI 6800 Ultra grading points you can see little growth, but also notice the same thing for the 7800 GTX in SLI. This either denotes the fact that SLI is adapted poorly in EQII or that nVidia SLI drivers at the time were in too bare an infancy to properly prorate "like they would" [were this to be the case] down the road. One other discrepancy to file is whether or not any of these marks had been CPU limited during testing. However unlikely -- construed from how low the FPS scores generally are [CPU limitation is pretty much iconic when benchmarks are bordering on "extremely over-the-top" {e.g. scenarios of 150-200+ FPS}] -- such a circumstance may be, it still is beheld as potentiality, IMO. But irrespective of two minute "counterpoints" [not even really so, accepting the overall ambiguousness behind such], the overwhelming points of interest substantiating the idea that EverQuest II's graphics architecture as being old outweigh the inconspicuous points of contest. With that, I think I can finally settle upon convincing even myself <img src="/smilies/69934afc394145350659cd7add244ca9.gif" border="0" alt="SMILEY" /> to suggest at least an assessment of the rendering schemes of EQII for a potential enhancement. - Seliri D. of Nagafen * EverQuest II settings descriptions of "Ultra High" and such associations are in reference to quality and not performance. (P.S. Your notion that networking infrastructure could stand as an inhibitor to the sanctity of graphical output does seem to be without much for buttressing. Multiple games inclusive of network facilitation, along with visuals rivaling -- if not equatable with -- ES4:O's receive higher benchmarks than EQII. Stutters strained out of issues of networking can be seen at all hardware levels and are especially distinguishable -- these aren't things you could run a muck with and end up befuddling yourself about, because impairments of networking are usually so distinct.)

Jynnan

05-10-2007, 09:46 AM

I run on Extreme settings, with Shadows on too, and mostly my FPS seem reasonable. I do get quite a bit of lag in Qeynos Harbour, but I'm usually just passing through, so it doesn't bother me much. The game looks incredible with Extreme settings turned on, compared with lower settings, almost a completely different experience! <img src="/smilies/3b63d1616c5dfcf29f8a7a031aaa7cad.gif" border="0" alt="SMILEY" />My rig comprises a Core2Duo 6600, twin 7950GT 512MB cards in SLi mode and 2 gig of RAM. I run in 1280x1024 Resolution (The max for my LCD screen).

Serso

05-10-2007, 09:50 AM

<cite>Jynnan wrote:</cite><blockquote>I run on Extreme settings, with Shadows on too, and mostly my FPS seem reasonable. I do get quite a bit of lag in Qeynos Harbour, but I'm usually just passing through, so it doesn't bother me much. The game looks incredible with Extreme settings turned on, compared with lower settings, almost a completely different experience! <img src="/smilies/3b63d1616c5dfcf29f8a7a031aaa7cad.gif" border="0" alt="SMILEY" />My rig comprises a Core2Duo 6600, twin 7950GT 512MB cards in SLi mode and 2 gig of RAM. I run in 1280x1024 Resolution (The max for my LCD screen).</blockquote> Everyone is saying things like "decent" or "reasonable" FPS -- do you have an actual estimation of average, if not the actual mean framerate? I s'pose it might help a little bit in acquainting with and discerning the response of EverQuest II toward some of the specifications running it. <img src="/smilies/ed515dbff23a0ee3241dcc0a601c9ed6.gif" border="0" alt="SMILEY" />

Of_mice_and_men

05-10-2007, 09:52 AM

Running high quality with shadows off on a system far worse than yours. You'll have no problem running this game with amazing graphics even if you don't play fully maxed out!Lavastorm is fine. Even with my old machine I didn;t have a problem there and certainly haven't noticed any issues there recently (spent quite a lot of time there in Jan/Feb 07)

Serso

05-10-2007, 09:56 AM

<cite>Of_mice_and_men wrote:</cite><blockquote>Running high quality with shadows off on a system far worse than yours. You'll have no problem running this game with amazing graphics even if you don't play fully maxed out!Lavastorm is fine. Even with my old machine I didn;t have a problem there and certainly haven't noticed any issues there recently (spent quite a lot of time there in Jan/Feb 07)</blockquote> Why did you cite Lavastorm specifically? I don't think there's a problem with peoples' ability to run the game in an adequate level of beauty...just the maximum level of beauty (where frailties of implementation are obviously more prone to transparency [conspicuousness]).

Despak

05-10-2007, 10:09 AM

<cite>Serso wrote:</cite><blockquote><cite>Jynnan wrote:</cite><blockquote>I run on Extreme settings, with Shadows on too, and mostly my FPS seem reasonable. I do get quite a bit of lag in Qeynos Harbour, but I'm usually just passing through, so it doesn't bother me much. The game looks incredible with Extreme settings turned on, compared with lower settings, almost a completely different experience! <img src="/smilies/3b63d1616c5dfcf29f8a7a031aaa7cad.gif" border="0" alt="SMILEY" />My rig comprises a Core2Duo 6600, twin 7950GT 512MB cards in SLi mode and 2 gig of RAM. I run in 1280x1024 Resolution (The max for my LCD screen).</blockquote> Everyone is saying things like "decent" or "reasonable" FPS -- do you have an actual estimation of average, if not the actual mean framerate? I s'pose it might help a little bit in acquainting with and discerning the response of EverQuest II toward some of the specifications running it. <img src="/smilies/ed515dbff23a0ee3241dcc0a601c9ed6.gif" border="0" alt="SMILEY" /></blockquote>Core 2 6400 @ 3.2ghz, 4 gb ddr2-800, 1 x 8800GTS 320mb, Vista Home Premium. Raiding Labs on Very High Quality with Particle effects turned down (to protect my eyes). Average FPS - 20-25. Peak @ 40+ lowest was 15fps.

Of_mice_and_men

05-10-2007, 10:12 AM

Because the OP mentioned Lavastorm in a later post.I can't answer his question fully not having his system. My system wont run extreme quality in anything I would call playable. I don't really feel the need to play at extreme was my point. It's pretty fantastic in High quality with shadows off.I know someone who can run the maxed out. Their machine is better than the OP's spec and they suffer no issues at all. Take that for what you will.

Norrsken

05-10-2007, 10:12 AM

You arent just getting it yet eh seliri? Game engines are built with different goals. Oblivion was built to be pretty. EQ2 was built as a massive networking app, with graphics. That networking engine (or whatever you want to call it) takes up processor cycles. And those cycles are also used for high level graphics manipulation, such as scene graphs, quad trees and such, things that gets done on the CPU rather than the graphics card. Having more to crowd in the CPU causes an FPS loss. I bet you that I can find several graphics engines that will whip oblivions [I cannot control my vocabulary], by just scaling off even more additional stuff, such as input handling and other similar things so that all the CPU does is handle the high level graphics. Heck, I've built game engines. <img src="/smilies/8a80c6485cd926be453217d59a84a888.gif" border="0" alt="SMILEY" /> Everything, and I mean EVERYTHING you do that takes up CPU time in games will have a small, sometimes not measurable effect, on the FPS. Unless you are running the graphics on hardware separated from the whatever else you may be doing elsewhere. Yes, there are games that may handle the networking better (I'd venture as far and say Eve online does), but then we have interactive objects in the world. Each and every object within your view frustum (or one of them, if several are employed) are probably checked to see if your mouse is hovering above it to know if that item is supposed to be highlighted as interative. That is not a very lightweight calculation. You could easily optimize this by not making items highligt, but only check for that mouse poiunter collision when you actually click. This is not how EQ2 does it however. And they had a reason not to.

Catsy

05-10-2007, 10:18 AM

FWIW, it's been my experience that EQ2 is an extremely inefficient, CPU-limited game that benefits very little from SLI and high-end video cards. Last year, my specs were: AMD Athlon 64 X2 4400+ Toledo 2.2Ghz ASUS A8N32 SLI Deluxe 2GB Corsair XMS PC 3200 2x BFGTech 7800 GT OC in SLI With this setup, I could run in balanced outdoors with particle settings and shadows/fauna turned down, and indoors somewhere between Very High and Max with particle effects turned down and shadows turned off. Disabling SLI had very little effect aside from needing to turn the texture qualities down due to cutting the video memory in half. Recently I upgraded to an EVGA 8800 GTX. Until the release of the Ultra last week, this was the single fastest video card model on the market. It runs everything I throw at it--FEAR, BF2, Oblivion, STALKER, Prey, Quake 4--in maximum quality settings at 1680x1050, with framerates smooth as glass. The only thing that chokes it is EQ2, which after doing some benchmarking over the last week is not appreciably faster. I can handle more particle effects and a little bit higher texture quality, but that's about it. Not being an EQ2 dev and having no direct knowledge of how they coded the game, the only thing I have to go on is a general understanding of the underlying technology, and an objective comparisons of performance. It makes sense that the code specific to a networked multiplayer game would impose some CPU overhead. We've seen this in nearly every other multiplayer game, where the single-player mode will always run a little faster than when you have, say, 32 players in the game. That explains /generally/ lower framerates on systems without substantial CPU horsepower. What it doesn't explain is the staggering performance hit EQ2 takes from ramping up the quality settings. It doesn't explain the difference in shadow rendering performance between, say, FEAR--which has to contend with adjusting hundreds of shadows in realtime based on complex physics model interactions--and EQ2, which starts choking even a relatively high-end video card like the 7-series when you turn on environmental shadows, which are largely static. It doesn't explain the difference in texture handling between, say, the Doom 3 engine--which at Ultra Quality loads all regular, normal map, diffuse map, etc textures completely uncompressed, consuming over 500M of texture memory--and EQ2, which struggles at its maximum quality settings and frankly doesn't have even close to the texture quality of the above mentioned games. I'm willing to bet that the performance Saavedra describes above owes far, far more to the Core Duo 2.6 than the 8800GTS. EQ2 is just a really inefficiently-coded game that depends too heavily on the CPU.

Serso

05-10-2007, 10:25 AM

<cite>Despak wrote:</cite><blockquote><cite>Serso wrote:</cite><blockquote><cite>Jynnan wrote:</cite><blockquote>I run on Extreme settings, with Shadows on too, and mostly my FPS seem reasonable. I do get quite a bit of lag in Qeynos Harbour, but I'm usually just passing through, so it doesn't bother me much. The game looks incredible with Extreme settings turned on, compared with lower settings, almost a completely different experience! <img src="/smilies/3b63d1616c5dfcf29f8a7a031aaa7cad.gif" border="0" alt="SMILEY" />My rig comprises a Core2Duo 6600, twin 7950GT 512MB cards in SLi mode and 2 gig of RAM. I run in 1280x1024 Resolution (The max for my LCD screen).</blockquote> Everyone is saying things like "decent" or "reasonable" FPS -- do you have an actual estimation of average, if not the actual mean framerate? I s'pose it might help a little bit in acquainting with and discerning the response of EverQuest II toward some of the specifications running it. <img src="/smilies/ed515dbff23a0ee3241dcc0a601c9ed6.gif" border="0" alt="SMILEY" /></blockquote>Core 2 6400 @ 3.2ghz, 4 gb ddr2-800, 1 x 8800GTS 320mb, Vista Home Premium. Raiding Labs on Very High Quality with Particle effects turned down (to protect my eyes). Average FPS - 20-25. Peak @ 40+ lowest was 15fps. </blockquote>Do you have all shadow and lighting optioning peaked? It's always interesting just to see how your rig performs on some titles, as levels of refined augmenting vary and, at least I feel, it's always good to know how some of the most up-to-date hardware reacts with some fairly timeworn graphics algorithms (e.g. to establish a feel for how well an exact program is taking advantage of the equipment relative to other circulating products).

Norrsken

05-10-2007, 10:32 AM

...

Serso

05-10-2007, 10:36 AM

Ulvhamne@Nagafen wrote: <blockquote>You arent just getting it yet eh seliri? Game engines are built with different goals. Oblivion was built to be pretty. EQ2 was built as a massive networking app, with graphics. That networking engine (or whatever you want to call it) takes up processor cycles. And those cycles are also used for high level graphics manipulation, such as scene graphs, quad trees and such, things that gets done on the CPU rather than the graphics card. Having more to crowd in the CPU causes an FPS loss. I bet you that I can find several graphics engines that will whip oblivions [I cannot control my vocabulary], by just scaling off even more additional stuff, such as input handling and other similar things so that all the CPU does is handle the high level graphics. Heck, I've built game engines. <img src="/smilies/8a80c6485cd926be453217d59a84a888.gif" border="0" alt="SMILEY" /> Everything, and I mean EVERYTHING you do that takes up CPU time in games will have a small, sometimes not measurable effect, on the FPS. Unless you are running the graphics on hardware separated from the whatever else you may be doing elsewhere. Yes, there are games that may handle the networking better (I'd venture as far and say Eve online does), but then we have interactive objects in the world. Each and every object within your view frustum (or one of them, if several are employed) are probably checked to see if your mouse is hovering above it to know if that item is supposed to be highlighted as interative. That is not a very lightweight calculation. You could easily optimize this by not making items highligt, but only check for that mouse poiunter collision when you actually click. This is not how EQ2 does it however. And they had a reason not to. </blockquote>Lol. I truly think you are presenting this factor in hyperbolic manner. With the processing solutions out today, many could laugh to go beyond 50% usage. However, whether or not "solutions" available are as effective as they could be (on terms of maximizing EQII's use of this power), I would say, is still up for deliberation AFAIK. Maybe in the past, but even still, I know this to not be the case because even on the lowest testing points, my system stands as a testament. I run an AMD Athlon 64 2.4 GHZ Clawhammer 4000 -- probably something of a close relative to the Athlon 64 4000 functioning in the article citing the benchmarks of the 6800 Ultra having 7 FPS with "Ultra High" video settings at 1028x768. The likely model of their AMD 4000 is the San Diego, the one with far less heat conduction and tenability for overclocking. And guess what? My CPU utilization never goes beyond the eighty-fifth and ninetieth percentile when dedicated on EQII. Restraint of framerate as a bi-product of one's CPU is what can be seen when the threshold of use reaches the one hundredth percentile. When that happens, you know that your benchmarking scores won't be as objective because you don't know how the application in question will react when it isn't pushing one of the components to an extreme end of the spectrum. Sotanyavejin@Guk wrote: <blockquote>FWIW, it's been my experience that EQ2 is an extremely inefficient, CPU-limited game that benefits very little from SLI and high-end video cards. Last year, my specs were: AMD Athlon 64 X2 4400+ Toledo 2.2Ghz ASUS A8N32 SLI Deluxe 2GB Corsair XMS PC 3200 2x BFGTech 7800 GT OC in SLI With this setup, I could run in balanced outdoors with particle settings and shadows/fauna turned down, and indoors somewhere between Very High and Max with particle effects turned down and shadows turned off. Disabling SLI had very little effect aside from needing to turn the texture qualities down due to cutting the video memory in half. Recently I upgraded to an EVGA 8800 GTX. Until the release of the Ultra last week, this was the single fastest video card model on the market. It runs everything I throw at it--FEAR, BF2, Oblivion, STALKER, Prey, Quake 4--in maximum quality settings at 1680x1050, with framerates smooth as glass. The only thing that chokes it is EQ2, which after doing some benchmarking over the last week is not appreciably faster. I can handle more particle effects and a little bit higher texture quality, but that's about it. Not being an EQ2 dev and having no direct knowledge of how they coded the game, the only thing I have to go on is a general understanding of the underlying technology, and an objective comparisons of performance. It makes sense that the code specific to a networked multiplayer game would impose some CPU overhead. We've seen this in nearly every other multiplayer game, where the single-player mode will always run a little faster than when you have, say, 32 players in the game. That explains /generally/ lower framerates on systems without substantial CPU horsepower. What it doesn't explain is the staggering performance hit EQ2 takes from ramping up the quality settings. It doesn't explain the difference in shadow rendering performance between, say, FEAR--which has to contend with adjusting hundreds of shadows in realtime based on complex physics model interactions--and EQ2, which starts choking even a relatively high-end video card like the 7-series when you turn on environmental shadows, which are largely static. It doesn't explain the difference in texture handling between, say, the Doom 3 engine--which at Ultra Quality loads all regular, normal map, diffuse map, etc textures completely uncompressed, consuming over 500M of texture memory--and EQ2, which struggles at its maximum quality settings and frankly doesn't have even close to the texture quality of the above mentioned games. I'm willing to bet that the performance Saavedra describes above owes far, far more to the Core Duo 2.6 than the 8800GTS. EQ2 is just a really inefficiently-coded game that depends too heavily on the CPU. </blockquote>I don't think that's the case. What it seems to be is that the intake and uptake of visual processes doesn't incorporate their output as efficaciously as could be. To me, this stands out as the only rational choice due to the spread of such broad sorts of brands, some with "CPU intensive networking" branches on par or exceeding that of EverQuest II.

Norrsken

05-10-2007, 10:41 AM

<cite>Serso wrote:</cite><blockquote>Ulvhamne@Nagafen wrote: <blockquote>You arent just getting it yet eh seliri? Game engines are built with different goals. Oblivion was built to be pretty. EQ2 was built as a massive networking app, with graphics. That networking engine (or whatever you want to call it) takes up processor cycles. And those cycles are also used for high level graphics manipulation, such as scene graphs, quad trees and such, things that gets done on the CPU rather than the graphics card. Having more to crowd in the CPU causes an FPS loss. I bet you that I can find several graphics engines that will whip oblivions [I cannot control my vocabulary], by just scaling off even more additional stuff, such as input handling and other similar things so that all the CPU does is handle the high level graphics. Heck, I've built game engines. <img src="/smilies/8a80c6485cd926be453217d59a84a888.gif" border="0" alt="SMILEY" /> Everything, and I mean EVERYTHING you do that takes up CPU time in games will have a small, sometimes not measurable effect, on the FPS. Unless you are running the graphics on hardware separated from the whatever else you may be doing elsewhere. Yes, there are games that may handle the networking better (I'd venture as far and say Eve online does), but then we have interactive objects in the world. Each and every object within your view frustum (or one of them, if several are employed) are probably checked to see if your mouse is hovering above it to know if that item is supposed to be highlighted as interative. That is not a very lightweight calculation. You could easily optimize this by not making items highligt, but only check for that mouse poiunter collision when you actually click. This is not how EQ2 does it however. And they had a reason not to. </blockquote>Lol. I truly think you are presenting this factor in hyperbolic manner. With the processing solutions out today, many could laugh to go beyond 50% usage. However, whether or not "solutions" available are as effective as they could be (on terms of maximizing EQII's use of this power), I would say, is still up for deliberation AFAIK. Maybe in the past, but even still, I know this to not be the case because even on the lowest testing points, my system stands as a testament. I run an AMD Athlon 64 2.4 GHZ Clawhammer 4000 -- probably something of a close relative to the Athlon 64 4000 functioning in the article citing the benchmarks of the 6800 Ultra having 7 FPS with "Ultra High" video settings at 1028x768. The likely model of their AMD 4000 is the San Diego, the one with far less heat conduction and tenability for overclocking. And guess what? My CPU utilization never goes beyond the eighty-fifth and ninetieth percentile when dedicated on EQII. Restraint of framerate as a bi-product of one's CPU is what can be seen when the threshold of use reaches the one hundredth percentile. When that happens, you know that your benchmarking scores won't be as objective because you don't know how the application in question will react when it isn't pushing one of the components to an extreme end of the spectrum. </blockquote>Are you watching the cpu load on the windows gauges? I dont trust them very much tbh. <img src="/smilies/8a80c6485cd926be453217d59a84a888.gif" border="0" alt="SMILEY" /> And then you have the overhead of context switches when doing different things. Page swapping and other neat stuff ya know. That the CPU isnt at 100% of cpu cycles doesnt mean its not at 100% capacity for the work its doing.

Serso

05-10-2007, 10:49 AM

Regardless, again, it is central to the premise that, in terms of benchmarking, the rigs at hand were on par with or better than mine. It's beyond the point whether your CPU is working at the most arduous level it can upon the tasks currently bestowed -- what is imperative here to the concept of scores being CPU-restricted is that the CPU reaches the largest possible workload it can handle. I've already mentioned why this is integral to the idea, so this highlight is all I can make note of.

Squigglle

05-10-2007, 12:26 PM

i got: duel hds 7200 rpm3200+ venice500 watt ps7600gt xxxnvidia chipsetddr400 1.5gigand i run it on medium, but they said its becuase my card isnt supported

Kodoku

05-10-2007, 12:54 PM

I'm running a core 2 duo e6600 with 2gb ddr2 800 ram and a 768mb XFX 8800 gtx oc (clocked at 1ghz) and I can run in extreme quality at 1920x1600 is most areas with very little graphics lag. There are exceptions of course, mostly in qeynos harbor, all of the people, particles, and reflections on the water usually end up killing performance from time to time. As to specific frame rates I coulden't tell you as I don't run anything that will give me an indicator.

Naubeta

05-10-2007, 12:59 PM

"Oblivion is a far better looking game than EQ2 at Extreme Quality" I disagree.

Catsy

05-10-2007, 01:12 PM

<cite>Serso wrote:</cite><blockquote>I don't think that's the case. What it seems to be is that the intake and uptake of visual processes doesn't incorporate their output as efficaciously as could be. To me, this stands out as the only rational choice due to the spread of such broad sorts of brands, some with "CPU intensive networking" branches on par or exceeding that of EverQuest II. </blockquote> The words are recognizably English, but I have no idea what you're trying to say here.

MrJekylls

05-10-2007, 01:17 PM

I have a laptop. Intel duo core 2 gig system with 2 gigs of ram and a 'go' video card (256 dedicated ram, 256 shared memory - 512 total); it has a 5400 rpm hard drive.I play on Very High quality with shadows turned off and a few other things turned up (I've increased the distance values to max for most things), and run at 1280x800 resolution. I get about 25-30 fps in most zones, which is very playable for me. I have no problem at all in EL.When I run on my external widescreen monitor at 1600x1200, I honestly don't get a very big performance hit. I think EQ2 isn't a very video card intensive game, and relies mainly on the cpu and memory. Any duo core and 2 gigs of RAM should eat up anything this game has to throw at it.

Kursa

05-10-2007, 01:34 PM

Ive got a quad core 4 gig RAM 8800 GTX vcard and you can run everything on max for the exception of shadows (which will take a hit). Other than that, it's good to go

Velsha

05-10-2007, 02:31 PM

Deathfyst Citadel nearly crippled my system on High Quality settings.

Morgane

05-10-2007, 02:59 PM

This seems like a good thread to ask: How does one find out what the fps rate is? I'd love to know what my fps rate is but don't have a clue how to find out. And I bet 90% of the people here know how to so help a girl out, would ya? <img src="/smilies/8a80c6485cd926be453217d59a84a888.gif" border="0" alt="SMILEY" />

Naubeta

05-10-2007, 03:49 PM

You can use something like FRAPS. One interesting thing is that your fps literally doubles when you turn off the UI in EQ2.

Generic123

05-10-2007, 04:11 PM

Sotanyavejin@Guk wrote: <blockquote>FWIW, it's been my experience that EQ2 is an extremely inefficient, CPU-limited game that benefits very little from SLI and high-end video cards. Last year, my specs were: AMD Athlon 64 X2 4400+ Toledo 2.2Ghz ASUS A8N32 SLI Deluxe 2GB Corsair XMS PC 3200 2x BFGTech 7800 GT OC in SLI With this setup, I could run in balanced outdoors with particle settings and shadows/fauna turned down, and indoors somewhere between Very High and Max with particle effects turned down and shadows turned off. Disabling SLI had very little effect aside from needing to turn the texture qualities down due to cutting the video memory in half. Recently I upgraded to an EVGA 8800 GTX. Until the release of the Ultra last week, this was the single fastest video card model on the market. It runs everything I throw at it--FEAR, BF2, Oblivion, STALKER, Prey, Quake 4--in maximum quality settings at 1680x1050, with framerates smooth as glass. The only thing that chokes it is EQ2, which after doing some benchmarking over the last week is not appreciably faster. I can handle more particle effects and a little bit higher texture quality, but that's about it. Not being an EQ2 dev and having no direct knowledge of how they coded the game, the only thing I have to go on is a general understanding of the underlying technology, and an objective comparisons of performance. It makes sense that the code specific to a networked multiplayer game would impose some CPU overhead. We've seen this in nearly every other multiplayer game, where the single-player mode will always run a little faster than when you have, say, 32 players in the game. That explains /generally/ lower framerates on systems without substantial CPU horsepower. What it doesn't explain is the staggering performance hit EQ2 takes from ramping up the quality settings. It doesn't explain the difference in shadow rendering performance between, say, FEAR--which has to contend with adjusting hundreds of shadows in realtime based on complex physics model interactions--and EQ2, which starts choking even a relatively high-end video card like the 7-series when you turn on environmental shadows, which are largely static. It doesn't explain the difference in texture handling between, say, the Doom 3 engine--which at Ultra Quality loads all regular, normal map, diffuse map, etc textures completely uncompressed, consuming over 500M of texture memory--and EQ2, which struggles at its maximum quality settings and frankly doesn't have even close to the texture quality of the above mentioned games. I'm willing to bet that the performance Saavedra describes above owes far, far more to the Core Duo 2.6 than the 8800GTS. EQ2 is just a really inefficiently-coded game that depends too heavily on the CPU. </blockquote>One of the biggest and most frequent complaints about DX9 is that it's a CPU pig. Yes there are games that deal with this better then EQ2, but none of them have there shear number and variety of objects they need to render. Just look at vanguard for example, it uses the normally smooth and reliable Unreal 2.5 engine, but when faced with the requirements of a large MMO it simple chokes. Would it have been better to try and do more with less the way LotRO has? Probably, but EQ2 is doing the things DX9 was supposed to be able to do. The fact that DX9 simply isn't efficient enough to pull this off in an MMO isn't knowledge that would have been available when EQ2 was in it's early design stages.

Naubeta

05-10-2007, 05:23 PM

Turbine have always made very good mmorpg engines, even if you didn't appreciate the games they built with them.

Despak

05-10-2007, 05:31 PM

<cite>Serso wrote:</cite><blockquote><cite>Despak wrote:</cite><blockquote><cite>Serso wrote:</cite><blockquote><cite>Jynnan wrote:</cite><blockquote>I run on Extreme settings, with Shadows on too, and mostly my FPS seem reasonable. I do get quite a bit of lag in Qeynos Harbour, but I'm usually just passing through, so it doesn't bother me much. The game looks incredible with Extreme settings turned on, compared with lower settings, almost a completely different experience! <img src="/smilies/3b63d1616c5dfcf29f8a7a031aaa7cad.gif" border="0" alt="SMILEY" />My rig comprises a Core2Duo 6600, twin 7950GT 512MB cards in SLi mode and 2 gig of RAM. I run in 1280x1024 Resolution (The max for my LCD screen).</blockquote> Everyone is saying things like "decent" or "reasonable" FPS -- do you have an actual estimation of average, if not the actual mean framerate? I s'pose it might help a little bit in acquainting with and discerning the response of EverQuest II toward some of the specifications running it. <img src="/smilies/ed515dbff23a0ee3241dcc0a601c9ed6.gif" border="0" alt="SMILEY" /></blockquote>Core 2 6400 @ 3.2ghz, 4 gb ddr2-800, 1 x 8800GTS 320mb, Vista Home Premium. Raiding Labs on Very High Quality with Particle effects turned down (to protect my eyes). Average FPS - 20-25. Peak @ 40+ lowest was 15fps. </blockquote>Do you have all shadow and lighting optioning peaked? It's always interesting just to see how your rig performs on some titles, as levels of refined augmenting vary and, at least I feel, it's always good to know how some of the most up-to-date hardware reacts with some fairly timeworn graphics algorithms (e.g. to establish a feel for how well an exact program is taking advantage of the equipment relative to other circulating products). </blockquote>I have lighting set to max, with 4 specular light. But don't bother with shadows as I think they are complete crap. Buggy, on sometimes, off others blah blah. If SoE ever manage to code some decent bug-free (ROFLMAO) shadows then I'll use them.

Boyar

05-10-2007, 09:35 PM

You should be fine with your setup. On a similar system(c2d 6600 2.4, 2GB, 8800GTX 768MB) I can run vh quality on my main monitor(1600x1200) and another instance with other account on my 1280x1024 secondary(more performance blended settings) monitor, all on the same video card.I only get performance issues in QH, and as I reach the kelethin platform level on busy evenings. I've never tried raiding multiclient, suspect that might be risky, but had no problems in Freethinkers last night at VH, and it was all very zippy and very pretty(lots of particles, we were mostly DPS). I'll have to try extreme now, I've always avoided it, as the folks that introduced me to EQ2 year before last warned me with the rumor that extreme quality and extreme performance were both a bit funky, so I've always favored settings between extremes, be it VH perf on my old system or VH quality on this one.Cheers, and welcome back

Miele

05-11-2007, 06:01 AM

20" widescreen for a 1650x1080 resolution AMD X2 3800+ 2gb ram semi fast SATA II HDD ATI 1950Pro 512 Mb ram I run at 25-35 fps with character shadows on, textures all to the max, 2 specualr lightining, 4 sources of light, flora maxed and maxed clip plane. What drags my fps down is Anti Aliasing at 4x and Anisotropic filter at 8x, but the game is so much more beatiful with them turned on. Environmental shadows are not an option, unless I'm soloing. Rarely my fps goes down in the high teens, it's a combination of lots of models displayed, lots of landscape objects and lots of particle effects to render.

IeU

06-23-2007, 03:56 PM

cmon ppl, post some high rez pic running under extreme setting

Tallisman

06-23-2007, 05:51 PM